Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

invalid username or password for correct password and worker process exits on signal 6 #15

Closed
gustopn opened this issue Jun 21, 2018 · 9 comments

Comments

@gustopn
Copy link

gustopn commented Jun 21, 2018

invalid username or password for...
worker process ... exited on signal 6
happens only when I am going over auth_digest.

@erikdubbelboer
Copy link

Is it possible for you to share the username and password so I can try to recreate the exact same situation? Also which browser and version of nginx are you using?

@gustopn
Copy link
Author

gustopn commented Jun 21, 2018

Yes, it does not matter what password you use, you might take test and test123.
The version of NGINX was the FreeBSD ports one, so nginx-1.14.0_4,2.
And the passwd file was created by the supplied script. However,
I am not suspecting the password to be false. It fails somewhere where it should give back to NGINX worker process, there it crashes. As far as I see the module seems to be quite old and NGINX changed in the meantime, so there may be an incompatibility there.

@erikdubbelboer
Copy link

I just compiled the module into nginx-1.14.0 for Ubuntu and everything works fine there. The nginx module system really hasn't changed so it's not an incompatibility there.

I don't have access to a FreeBSD machine I'm afraid so I can't test that. Maybe it's some modification that only the BSD version has.

Is there any way you could get a stacktrace or something?

@gustopn
Copy link
Author

gustopn commented Jun 23, 2018

Not today, but I may look into it sometime later and yes, that stacktrace idea would be a start.

@RvdE
Copy link

RvdE commented Sep 24, 2021

I have the same issue on FreeBSD 13.0-RELEASE.

Managed to get the following stacktrace:

Reading symbols from /usr/local/sbin/nginx...
[New LWP 1077745]
Core was generated by `nginx: worker process'.
Program terminated with signal SIGABRT, Aborted.
Sent by kill() from pid 82879 and user 80.
#0 kill () at kill.S:4
4 kill.S: No such file or directory.
(gdb) bt
#0 kill () at kill.S:4
#1 0x00000008008f79e0 in __fail (msg=) at /usr/src/lib/libc/secure/stack_protector.c:130
#2 0x00000008008f7950 in __stack_chk_fail () at /usr/src/lib/libc/secure/stack_protector.c:137
#3 0x000000080031ecc8 in ngx_http_auth_digest_evasion_tracking (r=r@entry=0x801b66050, alcf=, alcf@entry=0x8010f3af8, status=, status@entry=0)
at /wrkdirs/usr/ports/www/nginx/work/nginx-http-auth-digest-cd86418/ngx_http_auth_digest_module.c:1462
#4 0x000000080031d6e8 in ngx_http_auth_digest_handler (r=0x801b66050) at /wrkdirs/usr/ports/www/nginx/work/nginx-http-auth-digest-cd86418/ngx_http_auth_digest_module.c:289
#5 0x000000000027aa91 in ngx_http_core_access_phase (r=0x801b66050, ph=0x801b9b958) at src/http/ngx_http_core_module.c:1100
#6 0x000000000027a1a5 in ngx_http_core_run_phases (r=0x801b66050) at src/http/ngx_http_core_module.c:875
#7 ngx_http_handler (r=r@entry=0x801b66050) at src/http/ngx_http_core_module.c:858
#8 0x00000000002844aa in ngx_http_process_request (r=r@entry=0x801b66050) at src/http/ngx_http_request.c:2114
#9 0x00000000002b4100 in ngx_http_v2_run_request (r=0x801b66050) at src/http/v2/ngx_http_v2.c:3983
#10 0x00000000002b397a in ngx_http_v2_state_header_complete (h2c=0x801b19600, pos=0x801dd6785 "", end=0x801dd6785 "") at src/http/v2/ngx_http_v2.c:1916
#11 0x00000000002ae080 in ngx_http_v2_read_handler (rev=) at src/http/v2/ngx_http_v2.c:435
#12 0x000000000026c0c5 in ngx_kqueue_process_events (cycle=0x8010b0790, timer=, flags=1) at src/event/modules/ngx_kqueue_module.c:669
#13 0x0000000000261c8a in ngx_process_events_and_timers (cycle=cycle@entry=0x8010b0790) at src/event/ngx_event.c:247
#14 0x000000000026a71d in ngx_worker_process_cycle (cycle=0x8010b0790, data=) at src/os/unix/ngx_process_cycle.c:719
#15 0x0000000000268b7a in ngx_spawn_process (cycle=cycle@entry=0x8010b0790, proc=0x26a6b0 <ngx_worker_process_cycle>, data=0x6, name=0x21d015 "worker process", respawn=respawn@entry=7)
at src/os/unix/ngx_process.c:199
#16 0x00000000002698fa in ngx_reap_children (cycle=0x8010b0790) at src/os/unix/ngx_process_cycle.c:596
#17 ngx_master_process_cycle (cycle=cycle@entry=0x8010b0790) at src/os/unix/ngx_process_cycle.c:174
#18 0x0000000000240ff5 in main (argc=, argv=) at src/core/nginx.c:383

@erikdubbelboer
Copy link

Which version of nginx-http-auth-digest are you using? (there are no releases so which git hash?)
According to your stacktrace it crashed on:

node->drop_time = ngx_time();

But it's impossible for node to be NULL at this point. So I'm wondering if you're running an older version or if something else happened.

@RvdE
Copy link

RvdE commented Sep 24, 2021

I looked into the ports distinfo (which stores the filenames, checksums, etc). and I think it's hash cd86418.

I'll see what happens if I use the latest available one. To be continued.

@RvdE
Copy link

RvdE commented Sep 24, 2021

hmm... that's basically the latest version, since there's only 1 commit newer than that which changes some documentation.

The crash happened on previous versions of FreeBSD as well, so I'm not sure what's going on. Seems to be a FreeBSD only issue :(

@erikdubbelboer
Copy link

How are you installing nginx-http-auth-digest? I'm unfamiliar with FreeBSD so I'm curious how a module like this is installed.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

3 participants